The new product provides cost certainty, as customers will know upfront the duration of the job, the number of GPUs they will use, and the total cost. The service displays the total cost for the timeframe and resources, which users can adjust according to their budget and resource needs before purchasing. The feature is now generally available in the AWS US East (Ohio) region.
Key takeaways:
- AWS has launched Amazon Elastic Compute Cloud (EC2) Capacity Blocks for ML to allow customers to rent GPUs for a defined amount of time.
- The product provides access to NVIDIA H100 Tensor Core GPUs instances in cluster sizes of one to 64 instances with 8 GPUs per instance.
- Customers can reserve time for up to 14 days in 1-day increments, up to 8 weeks in advance, and the instances will shut down automatically when the timeframe is over.
- The new feature is generally available starting today in the AWS US East (Ohio) region.